1. PASQUALE BLASI INAF/Arcetri Astrophysical Observatory, Firenze & GSSI/Gran Sasso Science Institute, L’Aquila Napoli, June 19 2014

2. DO WE HAVE A STANDARD MODEL? THE MAIN GUIDANCE FROM DATA HAS COME FROM: - PROTON KNEE AT ~ THE ALL-PARTICLE KNEE - B/C RATIO (and other secondary/primary ratios) THE MOST ‘STANDARD’ MODEL WE HAVE CONSISTS OF TWO PIECES: 1.Particles propagate diffusively in the Galaxy, which should explain the secondary/primary ratios and unstable isotopes + diffuse bkgnds. 1.Particles are accelerated in supernova remnants (SNRs) through Diffusive Shock Acceleration

3. SPECTRA ALL PARTICLE SPECTRUM PROTONS P+He (KASCADE) KNEE TRANSITION ? GZK or SOMETHING ELSE?

4. B/C RATIO THE DECREASE OF B/C WITH Energy/nucleon IS THE BEST SIGN OF A RIGIDITY DEPENDENT GRAMMAGE TRAVERSED BY COSMIC RAYS ON THEIR WAY OUT OF THE GALAXY For relativistic E

5. HOW STANDARD IS THE STANDARD MODEL OF TRANSPORT? EVEN COMPLEX COMPUTATION CODES SUCH AS GALPROP ADOPT IMPORTANT SIMPLIFICATIONS… RELAXING THESE ASSUMPTIONS  CORRECTIONS AT THE ZERO ORDER SEVERAL COMPLICATIONS: 1.DIFFUSION COEFFICIENT NON SPATIALLY CONSTANT 2.ANISOTROPIC DIFFUSION (PARALLEL vs PERPENDICULAR) 3.EFFECT OF SELF-GENERATION OF WAVES INDUCED BY CR 4.DAMPING OF WAVES AND ITS EFFECTS ON CR PROPAGATION 5.CR INDUCED GALACTIC WINDS 6.CASCADING OF MODES IN WAVENUMBER SPACE EACH ONE OF THESE PHYSICAL MECHANISMS CHANGES THE PREDICTED SPECTRA AND ANISOTROPIES IN A SUBSTANTIAL WAY

6. NON SEPARABLE D(E,z) THE STANDARD RULE OF THUMB THAT n(E)~Q(E) t (E) ~ Q(E)/D(E)~E -g-d IS ONLY VALID FOR SPATIALLY CONSTANT DIFFUSION OR FOR SEPARABLE D(E,z)=F(E)G(z) EASIEST INSTANCE OF NON-SEPARABILITY: H2H1H2H1 H2H2 H1H1 E D(E) D1D1 D2D2 D1D1 D2D2

7. NON SEPARABLE D(E,z) THE SIMPLEST FORM OF THE TRANSPORT EQUATION CAN BE WRITTEN AS: WHERE THE CR DENSITY IN THE DISC IS ( e.g. Dogiel 2001, Tomassetti 2012 ) : E D(E) D1D1 D2D2 E f(E)

8. ANISOTROPIC DIFFUSION EVEN FOR ISOTROPIC TURBULENCE, DIFFUSION IS NOT ISOTROPIC De Marco, PB, Stanev 2007 ISOTROPIC TURBULENCE PARALLEL PERPENDICULAR De Marco +, 2007 NUMERICAL SIMULATIONS OF PARTICLE PROPAGATION IN ISOTROPIC TURBULENCE SHOW THAT 1) EVEN FOR d B/B~1, D par > D perp 2) PAR AND PERP D(E) HAVE DIFFERENT ENERGY DEPENDENCES

9. AN OLD PROBLEM … Ion-Neutral Damping Kulsrud & Cesarsky (1971) ION-NEUTRAL DAMPING IS EFFECTIVE FOR LARGE WAVENUMBER NAMELY FOR LOW PARTICLE MOMENTUM (IN TERMS OF RESONANT SCATTERING). FOR V A =20 km/s and n i <<n H : EFFECTIVE SPATIAL SEGREGATION (MOLECULAR CLOUDS) IS REQUIRED IN ORDER TO ALLOW FOR DIFFUSION IN THE GALACTIC DISC

10. CR INDUCED SCATTERING ENSAMBLES OF CHARGED PARTICLES WITH SUPERALFVENIC VELOCITY INDUCE STREAMING INSTABILITY ON k ~ 1/LARMOR RADIUS IN THE CONTEXT OF CR PROPAGATION IN THE GALAXY, THIS WAS FIRST DISCUSSED BY Skilling (1975). Holmes (1975) DISCUSSED THE RELATIVE IMPORTANCE OF GROWTH AND DAMPING OF THE WAVES RESPONSIBLE FOR CR SCATTERING COSMIC RAYS MAY PLAY AN ACTIVE ROLE IN THEIR OWN DIFFUSIVE PROPAGATION

11. THE ROLE OF CR ON THEIR OWN TRANSPORT THE GROWTH RATE OF THE UNSTABLE MODES INDUCED BY CR IS PROPORTIONAL TO THEIR SPATIAL GRADIENT WHERE THE DISTRIBUTION FUNCTION OF CR IS FOUND FROM THE DIFFUSION EQUATION WITH:

12. CR TRANSPORT IN SELF-GENERATED WAVES a DENOTES THE TYPE OF NUCLEUS (BOTH PRIMARIES AND SECONDARIES ARE INCLUDED) PB, Amato & Serpico 2012, Aloisio & PB 2013 TRANSPORT EQUATION FOR WAVES

13. GENERAL TRENDS, D(E), X(E) LOW ENERGY HARDENING HIGH ENERGY HARDENING Aloisio & PB 2013 ADVECTION DIFFUSION Aloisio & PB 2013 DIFFUSION COEFFICIENT GRAMMAGE

14. MAIN IMPLICATIONS Aloisio & PB 2013 B/C C/O

15. NO BREAKS IN AMS-02 1.APPARENTLY NO SIGN OF SPECTRAL BREAKS 2.OVERALL SPECTRAL STRUCTURE RATHER DIFFERENT 3.ALL AGREE THAT He SPECTRUM IS HARDER THAN H SPECTRUM 4.NORMALIZATION OF FLUXES IN GOOD AGREEMENT WITH CREAM

16. HOW STANDARD IS THE STANDARD MODEL OF CR ACCELERATION? CR ACCELERATION EFFICIENCIES OF 10-20%  NON LINEAR DSA MAXIMUM ENERGIES  MAGNETIC FIELD AMPLIFICATION, MOST LIKELY DUE TO ACCELERATED PARTICLES (ADDITIONAL NON LINEARITY) SPECTRUM AT THE EARTH  A RATHER COMPLEX SUPERPOSITION OF PARTICLES ADVECTED DOWNSTREAM AND ESCAPING FLUX BUT THERE ARE NUMEROUS ASPECTS OF DSA THAT GO WAY BEYOND THIS QUASI-STANDARD PICTURE…

17. ACCELERATION AT SNR SHOCKS MASS MOMENTUM ENERGY CR TRANSPORT TWO MAIN IMPLICATIONS: 1.CONCAVE SPECTRA 2.MODIFIED SHOCK HEATING

18. EFFECTS OF NLDSA PB, Gabici & Vannoni, 2005 THERMAL EFFECTS CONCAVITY

19. PROBLEMS WITH PREDICTED SPECTRA The non linear theory of DSA (as well as the test particle theory) all predict CR spectra close to E -2 and even harder than E -2 at E>10 GeV This finding does not sit well with: 1)CR ANISOTROPY (THE REQUIRED D(E) HAS TO SCALE AS E 0.75 ) 2)GAMMA RAY SPECTRA FROM SELECTED SNRS Caprioli 2011 Amato & PB 2012

20. POSSIBLE REASONS FOR STEEPER SPECTRA 1.DYNAMICAL REACTION OF AMPLIFIED MAGNETIC FIELD (Caprioli +, 2008,2009) 2. SHOCK MODIFICATION INDUCED BY NEUTRALS IN THE SHOCK REGION (PB+, 2012) SEE TALK BY G. MORLINO AND DISCUSSION LATER IN THIS TALK 3. FINITE VELOCITY OF SCATTERING CENTERS (Caprioli+, 2010, Zirakashvili, Ptuskin, Seo 2010) IF MAGNETIC FIELD AMPLIFICATION LEADS TO FAST MOVING WAVES IN THE PLASMA FRAME, THEN THE COMPRESSION FACTOR RELEVANT FOR ACCELERATION IS THAT OF THE SCATTERING CENTER’S VELOCITY. THIS EFFECT MAY LEAD TO STEEPENING AS WELL AS TO HARDENING DEPENDING ON HELICITY. CHANGE IN THE EQUATION OF STATE  LESS COMPRESSION @ SHOCK  STEEPER SPECTRA

21. HARD He SPECTRUM THE ONLY VIABLE EXPLANATION OF THE HARDER He SPECTRUM SO FAR IS BASED ON THE INJECTION MODEL OF Malkov et al. 2012. THE PREFERENTIAL INJECTION OF He 2+ IS DUE TO ITS LARGER LARMOR RADIUS, BUT THEN IT IS REQUIRED TO DECREASE WITH M A INDEPENDENT OF THIS MODEL PROVIDING OR NOT THE CORRECT INTERPRETATION OF THE DATA, IT GIVES US A FLAVOR OF HOW MUCH OUR ‘STANDARD MODEL’ IS SENSITIVE TO POORLY KNOWN ASPECTS OF THE MICROPHYSICS THIS IS, IN A WAY, A RECENT FINDING, DUE TO THE FACT THAT THE EFFECTS WE ARE SEARCHING FOR ARE AT THE SAME LEVEL OF EXPERIMENTAL SYSTEMATICS…

22. MAGNETIC FIELD AMPLIFICATION TYPICAL SIZE OF THE FILAMENTS ~ 10 -2 parsec The emission in the filaments is non-thermal synchrotron of the highest energy electrons in the accelerator Comparison with the observed thickness leads to an estimate for the local field

23. MAGNETIC FIELD AMPLIFICATION AND P MAX THE EVIDENCE FOR B-FIELD AMPLIFICATION IS CRUCIAL FOR AT LEAST TWO REASONS:  IN THE ABSENCE OF THIS PHENOMENON THE MAXIMUM ENERGY OF ACCELERATED PARTICLES IS ~1-10 GeV ONLY  MAGNETIC FIELD AMPLIFICATION HAS LONG BEEN EXPECTED AS A BY- PRODUCT OF CR ACCELERATION (STREAMING INSTABILITY) IN THE BOHM ASSUMPTION FOR THE TURBULENT COMPONENT: CLOSE TO THE TIME OF BEGINNING OF THE SEDOV PHASE…

24. THE QUESTION IS ‘WHETHER THE CR INDUCED INSTABILITIES GROW FAST ENOUGH TO ACHIEVE THE GOAL A RELEVANT QUANTITY IS WEAK CR INDUCED MODIFICATION STRONG CR INDUCED MODIFICATION Alfven waves are excited, with a small imaginary part of frequency. The growth rate reads: The resonant nature of the instability limits it to d B/B~1 Quasi-purely growing non resonant modes are excited (Bell 2004) together with aperiodic resonant modes with Im[ w ]~Re[ w ]. In the absence of additional dynamics the non resonant modes do not scatter CR effectively [k>>1/r L (p)]

25. STRONGLY CR MODIFIED REGIME THE MAX GROWTH RATE IS: WHERE J CR IS THE CR CURRENT. IN THE PICTURE OF Schure, Bell, Reville THE RELEVANT CURRENT IS THAT OF PARTICLES THAT ATTEMPT TO ESCAPE THE ACCELERATOR AT ENERGY E THE d B GROWS BUT ALSO CHANGES SCALE (PLASMA IS MOVED AROUND WITH FORCE ~(1/c) J CR d B). SATURATION OCCURS WHEN

26. MAXIMUM ENERGY FOR A SNR IN THE ISM THE SATURATION IS TYPICALLY REACHED WHEN g W,max t ~ 5, WHICH IMPLIES, EVALUATED AT THE BEGINNING OF THE SEDOV PHASE: THIS VALUE OF A FEW HUNDRED TeV IS RATHER WEAKLY DEPENDENT UPON THE VALUES OF PARAMETERS IT APPEARS UNLIKELY THAT PROTONS MAY BE ACCELERATED UP TO THE KNEE IN A SN TYPE Ia, ALTHOUGH THE PROBLEM IS BY A FACTOR ~10 Schure et al. 2012

27. MAXIMUM ENERGY FOR A SNR IN A SUPERGIANT WIND CORE COLLAPSE SN OFTEN EXPLODE IN THE WIND OF THE GIANT PROGENITOR. THE GAS DENSITY IN THE WIND IS THE MAX ENERGY OF ACCELERATED PARTICLES IS IN THE DENSE WIND THE SEDOV PHASE IS REACHED AT DISTANCE CHOOSING: AND WITH A SHOCK SPEED 20000 km/s close to the knee… STILL, PARAMETERS NEED TO BE PUSHED TO EXTREME VALUES

28. TYCHO Morlino & Caprioli 2011 STEEP SPECTRUM HARD TO EXPLAIN WITH LEPTONS E max ~500 TeV THE STEEP SPECTRUM MAY BE THE RESULT OF FINITE VELOCITY OF SCATTERING CENTERS ( Caprioli et al. 2010, Ptuskin et al. 2010, Morlino & Caprioli 2011 ) OR MAY RESULT FROM AN INHOMOGENEOUS MEDIUM ( Berezhko et al. 2013 )

29. ALL PARTICLE SPECTRUM P+He SPECTRUM

30. WHERE DO GALACTIC CRs END? A SUPERPOSITION OF EXPONENTIAL CUTOFFS SCALING WITH Z FROM THE KNEE LEADS TO A DEFICIT AT E>10 7 GeV KASCADE GRANDE DATA SHOW A RATHER COMPLEX IMPLEMENTATION OF THE TRANSITION REGION, WHICH IS NOT EASY TO RECONCILE WITH OBSERVATIONS OF UHECR SPECTRA AND MASS COMPOSITION Amato & PB 2012 Apel et al. 2013

31. A LOOK FROM THE OTHER END… DIP THE DIP IS THE ONLY EXPLANATION OF THE TRANSITION THAT IS COMPATIBLE WITH THE BASIC PREDICTIONS OF THE SNR PARADIGM FOR GALACTIC COSMIC RAYS BUT IT IS INCOMPATIBLE WITH THE OBSERVED COMPOSITION

32. PHOTODISINTEGRATION OF NUCLEI FOR NUCLEI THE MOST IMPORTANT PHYSICAL PROCESS IS THE PHOTO- DISINTEGRATION OF THE NUCLEI IN SCATTERING WITH PHOTONS OF THE EBL+CMB INTERESTINGLY, THE FACT THAT THE BINDING ENERGY OF A NUCLEON IN A NUCLEUS DOES NOT DEPEND DRAMATICALLY UPON THE NUCLEUS, LEADS TO A WEAK DEPENDENCE OF THE CRITICAL LORENTZ FACTOR:

33. Aloisio et al. 2013

34. TRANSITION AND UHECR VERY HARD INJECTION SPECTRA REQUIRED IF SOURCES GENERATE NUCLEI Aloisio, Berezinsky & PB 2014

35. TRANSITION AND UHECR Aloisio, Berezinsky & PB 2014

36. TRANSITION AND UHECR Additional Galactic Component Aloisio, Berezinsky & PB 2014 IF THE REQUIRED ADDITIONAL COMPONENT IS GALACTIC IN ORIGIN, IT HAS TO BE LIGHT. BUT THIS IS NOT CONSISTENT WITH THE ANISOTROPY MEASURED BY AUGER AT 10 18 eV IF THE REQUIRED ADDITIONAL COMPONENT IS GALACTIC IN ORIGIN, IT HAS TO BE LIGHT. BUT THIS IS NOT CONSISTENT WITH THE ANISOTROPY MEASURED BY AUGER AT 10 18 eV

37. TRANSITION AND UHECR Additional Extragalactic Component Aloisio, Berezinsky & PB 2014

38. TRANSITION AND UHECR Additional Extragalactic Component Aloisio, Berezinsky & PB 2014

39. BALMER LINES THIS FIELD IS STILL YOUNG, BUT IT HAS A HUGE POTENTIAL IN TERMS OF ‘MEASURING’ THE AMOUNT OF COSMIC RAYS CLOSE TO SNR SHOCKS THE PHYSICS QUESTION IS: “WHAT HAPPENS WHEN A COLLISIONLESS SHOCK CROSSES A PARTIALLY IONIZED MEDIUM?” THE WIDTH OF THE BALMER LINE(S) PROVIDES POWERFUL INFORMATION ON THE IONS TEMPERATURE DOWNSTREAM OF THE SHOCK  CR CALORIMETRY !!! SHOCK VELOCITY INFLOWING NEUTRALS AND IONS DvDv NEUTRALS IONS

40. BROAD AND NARROW H a LINE TWO PROCESSES GET ACTIVATED: CHARGE EXCHANGE IONIZATION OF NEUTRAL HYDROGEN + + Hot ion Cold neutralhot neutral Cold ion n=1 n=2 BALMER LINE FROM ATOMS WITH CE DOWNSTREAM H a REFLECTS TEMPERATURE OF IONS DOWNSTREAM BROAD BALMER LINE BALMER LINE FROM ATOMS WITH NO CE DOWNSTREAM H a REFLECTS TEMPERATURE UPSTREAM NARROW BALMER LINE

41. ANOMALOUS H a LINE WIDTHS IN THE PRESENCE OF PARTICLE ACCELERATION TWO THINGS HAPPEN: LOWER TEMPERATURE DOWNSTREAM A PRECURSOR APPEARS UPSTREAM NEUTRALS IONS BROAD BALMER LINE GETS NARROWER NARROW BALMER LINE GETS BROADER BROAD BALMER LINE GETS NARROWER NARROW BALMER LINE GETS BROADER DvDv

42. MATHEMATICAL APPROACH IONS ARE TREATED AS A PLASMA WITH GIVEN DENSITY AND A THERMAL DISTRIBUTION NEUTRAL ATOMS ARE DESCRIBED USING A BOLTZMAN EQUATION WITH SCATTERING TERMS DESCRIBING CHARGE EXCHANGE AND IONIZATION PB, Morlino, Bandiera, Amato & Caprioli, 2012

43. MATHEMATICAL APPROACH PB, Morlino, Bandiera, Amato & Caprioli, 2012 IONS AND NEUTRALS ARE CROSS-REGULATED THROUGH MASS, MOMENTUM AND ENERGY CONSERVATION:

44. NEUTRAL RETURN FLUX SHOCK VELOCITY NEUTRALS AND IONS 1 v/(Vs/4) V perp =0 V<0 A NEUTRAL ATOM CAN CHARGE EXCHANGE WITH AN ION WITH V<0, THEREBY GIVING RISE TO A NEUTRAL WHICH IS NOW FREE TO RETURN UPSTREAM THIS NEUTRAL RETURN FLUX LEADS TO ENERGY AND MOMENTUM DEPOSITION UPSTREAM OF THE SHOCK! PB et al. 2012

45. DISTRIBUTION FUNCTIONS IN PHASE SPACE THE DISTRIBUTION FUNCTIONS OF NEUTRALS ARE NOT MAXWELLIAN IN SHAPE BUT APPROACH SUCH SHAPE AT DOWNSTREAM INFINITY NEUTRAL RETURN FLUX PB+ 2012

46. NEUTRAL INDUCED PRECURSOR EVEN FOR A STRONG SHOCK (M>>1) THE EFFECTIVE MACH NUMBER OF THE PLASMA IS DRAMATICALLY REDUCED DUE TO THE ACTION OF THE NEUTRAL RETURN FLUX PB+ 2012

47. ACCELERATION OF TEST PARTICLES PB+ 2012

48. NON LINEAR CR ACCELERATION IN PARTIALLY IONIZED PLASMAS BOLTZMANN EQUATION FOR NEUTRALS CR TRANSPORT EQUATION GENERALIZED CONSERVATION EQUATIONS TRANSPORT OF WAVES

49. Balmer Line Zoom on the narrow Balmer line HEATING IN THE PRECURSOR SHAPE OF THE BALMER LINE Morlino +, 2012

50. SNR 0509-67.5 Helder et al. 2009 Morlino et al. 2013 NE SW SHOCK VELOCITY RATHER UNCERTAIN DISTANCE WELL KNOWN (LMC): 50±1 kpc

51. SNR 0509-67.5 Morlino et al. 2013 FOR SHOCK VELOCITY ~5000 km/s A LOWER LIMIT OF 5-10% TO THE CR ACCELERATION EFFICIENCY CAN BE IMPOSED

52. RCW 86 Balmer line + proper motion Thermal X-ray spectrum DISTANCE TO THIS SNR RATHER UNCERTAIN WITH VALUES RANGIND FROM 2 TO 3 kpc, WITH MOST LIKELY VALUE OF 2.5 kpc IN THE ABSENCE OF AN INDEPENDENT INFORMATION ON THE ELECTRON-ION EQUILIBRATION, THE BALMER LINE WIDTH IS COMPATIBLE WITH NO CR ACCELERATION IN 3 REGIONS HOWEVER THERE ARE X-RAY MEASUREMENTS OF THE ELECTRON TEMPERATURE Helder, Vink and Bassa 2011

53. RCW 86 IF THE MEASURED ELECTRON TEMPERATURE IS THE ACTUAL T e DOWSNTREAM, THEN ALL MEASURED FWHM OF THE BROAD BALMER LINE SUGGEST EFFICIENT CR ACCELERATION

54. RCW 86 – FILAMENT SE OUT CASE A CASE B 10% neutrals 50% neutrals A NON THERMAL PRESSURE OF ABOUT 20-30% IS REQUIRED TO EXPLAIN AT THE SAME TIME THE FWHM OF THE BALMER LINE AND THE VALUE OF T e

55. RCW 86 – FILAMENT SE OUT : a case of return flux HIGH NON THERMAL PRESSURE DOES NOT NECESSARILY IMPLY A LARGE FLUX OF HIGH ENERGY COSMIC RAYS… INTERMEDIATE BALMER LINE

56. SUMMARY 1.THEORETICAL ISSUES AND SUCCESSES 2. OBSERVATIONAL ISSUES AND SUCCESSES MAXIMUM ENERGY AND MAGNETIC FIELD AMPLIFICATION ESCAPE OF CR FROM THEIR SOURCES INJECTION SPECTRA AND ANISOTROPY PROPAGATION IN THE GALAXY (DIFFUSION, WINDS, SELF-GENERATED WAVES VS PRE- EXISTING TURBULENCE, CASCADING TRANSITION TO EXTRAGALACTIC CR POTENTIAL FOR DISCOVERY IN BALMER LINE OBSEVRATIONS OF SNR – A LOT PHYSICS INVOLVED POSITION OF THE KNEE IN THE LIGHT COMPONENT STILL UNCLEAR STRONG DEPENDENCE OF RESULTS ON MC OF SHOWERS (KNEE AND ABOVE) SPECTRAL BREAKS POTENTIALLY INTERESTING BUT NOT CONFIRMED (PAMELA vs AMS) – ROLE OF SYSTEMATICS B/C STILL NOT CONCLUSIVE IN NAILING DOWN D(E)